1. Technical Field
The present invention relates to a biosensor used to detect or quantify a specific component in a liquid sample, and more particularly relates to a biosensor capable of accurate measurement with excellent reproducibility, and to a method of producing this biosensor, and a detection system in which this biosensor is used.
2. Description of the Prior Art
A biosensor used to measure blood glucose levels, which is a typical biosensor, can easily quantify the glucose in blood by mainly utilizing an electrochemical reaction, and is made up of two or more electrodes, a reagent that reacts with glucose, an electron transfer substance that allows the reaction to proceed smoothly, and so forth.
The technological trend in recent years with biosensors used to measure blood glucose levels is that the time it takes for measurement is being shortened to make the devices easier for patients to use, and that the amount of blood required for measurement is being reduced in order to minimize pain during puncture. In fact, the latest technology makes possible biosensors whose measurement time is only 5 seconds and in which the amount of blood measured is 0.3 μL.
Meanwhile, the rise in treatment costs that accompanies the global increase in diabetes patients in recent years has led to a greater quantity of biosensors used for measuring blood glucose levels and to a reduction in price. As a result, though, of the biosensors sold for measuring blood glucose levels, some sacrifice performance for a lower price. And measurement performance is also sometimes sacrificed for the sake of a shorter measurement time and smaller amount of blood.
Because of the above situation, there is currently a need in the marketplace for an improved and inexpensive biosensor used to measure blood glucose levels that does not sacrifice measurement performance, is easier for diabetes patients to use, and inflicts less pain.
There are known biosensors for quantifying a specific component in a liquid sample, but technical advances and terminology used by inventors vary, and can be extremely hard to understand, so the same terminology will be used, while not departing from the original meaning, to make the differences from prior art as clear as possible.
Biosensor technology has a long history, and can be traced all the way back to 1956, which Leland C. Clark suggested that an oxygen electrode capable of measuring the oxygen in a solution could be applied to a biosensor (see Patent Document 1 and Non-Patent Document 1).
Then, in 1962, Leland C. Clark described that he was able to produce an amperometric glucose oxygen electrode by using a dialysis membrane to seal a glucose oxidation enzyme in a Clark oxygen electrode (see Non-Patent Document 2).
In 1967, Updike and Hicks published an “enzyme electrode” with which a substrate in a solution could actually be measured by fixing an oxidation enzyme to an oxygen electrode surface with polyacrylamide gel, so that the oxygen was proportionally consumed in a substrate oxidation reaction within this fixing layer, and the measured current value decreased (see Patent Document 3 and Non-Patent Document 4).
The above-mentioned Clark oxygen electrode is usually a two-electrode electrolytic cell that has a platinum working electrode and a silver/silver chloride reference electrode (counter electrode). The following are the two greatest technical advances provided by a Clark oxygen electrode. 1) Because a gas permeable membrane is affixed over a platinum electrode, the measurement system and the electrode system are separate, so current response is stable and there is no admixture of substances that would hinder measurement, and 2) the oxygen consumption type of biorecognition ability of fixed oxidation enzymes, or mitochondria and other organellas, or aerobic cells (such as microbes), is utilized, and the dissolved oxygen in a liquid sample that has been reduced as a result of a substrate specific reaction is quantified, which allows a certain substrate to be measured (see, for example, Patent Documents 2, 4, 5, and 8, and Non-Patent Documents 3 and 6).
However, a drawback to a biosensor that makes use of a Clark oxygen electrode is that the measurable range is limited by the amount of oxygen dissolved in the liquid sample. In view of this, in 1975 Thomas et al. employed a coenzyme oxidized nicotinamide adenine dinucleotide (hereinafter referred to as NAD) as an electron transfer substance between an enzyme and an electrode, and reported that this was extremely effective in terms of overcoming the above-mentioned drawback to the measurable range of a Clark oxygen electrode. According to this invention, the active level of lactate dehydrogenase (hereinafter referred to as LDH) was proven to have a correlation with the result for oxidation current of a coenzyme reducted nicotinamide adenine dinucleotide (hereinafter referred to as NADH). What produces this correlation is that NADH increases in proportion to substrate concentration when a total of three electrodes are used (an auxiliary and two reference electrodes) to apply a constant voltage with a potentiostat, so the way of being converted from NADH to NAD is seen as a result of an electrochemical reaction (see Non-Patent Document 7).
Also, biosensors to which the above-mentioned invention is applied have undergone further research and development to make them more practical, such as fixing an enzyme and a coenzyme (see Patent Documents 11 and 13, for example).
Part of the above-mentioned drawback to biosensors involving the use of a Clark oxygen electrode, that the measurement range was limited, was solved by a method in which hydrogen peroxide produced in a reaction by an oxidation enzyme was amperometrically detected. This is because the amount of hydrogen peroxide increases as the enzyme reaction proceeds, unlike methods in which oxygen is measured. However, even with a method in which hydrogen peroxide is measured, a drawback is that if the liquid sample is blood, depending on the applied voltage (near +0.7 V to Ag/AgCl), various interference substances in the blood, such as ascorbic acid or uric acid, may affect the measurement result depending on their concentration.
Newman et al. introduced, as a method for eliminating the effect of the above-mentioned interference substances in amperometric detection, a method in which just hydrogen peroxide is selectively transmitted to the electrode through a cellulose membrane or the like, and just hydrogen peroxide is measured amperometrically (see Patent Documents 9 and 10).
Upon receiving these research results, the Yellow Spring Instrument company (Ohio, United States of America) in 1975 marketed a large glucose biosensor with which hydrogen peroxide was selectively transmitted with a cellulose membrane or the like, and hydrogen peroxide was measured amperometrically.
Although not currently commercially available, in 1976 the Miles company (Illinois, United States of America) marketed under the trade name of Biostator a large artificial pancreas for bedside use that employed a glucose biosensor developed by Clemens et al. (see Patent Document 12).
Thus, biosensors featuring oxygen electrodes, which were developed by Leland C. Clark and are now called first-generation biosensors, have been the subject of continuing research by fixing various oxidation enzymes. The subject of this research has been greatly expanded from just the medical field to include the environmental field and the food industry, and some of these research projects have seen commercial application, albeit in small numbers.
Even today, however, the above-mentioned industries and products have yet to achieve major commercial success. A reason for this is that, in addition to the technical problems listed up to now, first-generation biosensors are produced by glass working, which does not lend itself to mass production, and when high manufacturing costs, as well as the production of gas permeable membranes and vessels for holding electrolyte are taken into account, the sensor ends up being so large in size that it is inconvenient to use.
In view of this, in an effort to solve manufacturing problems, in 1990 Karube et al. invented a microbiosensor in which silicon was used instead of glass for the base plate, a non-liquid aqueous electrolyte-containing material was put in a groove formed by anisotropic etching, and a gas permeable membrane was added (see Patent Document 23).
However, although inventions such as the one above were intended to overcome the drawbacks to first-generation biosensors, the biosensors to which these inventions were applied have yet to reach the market. This is probably because the problems of manufacturing cost and ease of use have not been solved.
In 1992, Kawaguri et al. and Nankai et al. reported on a technique for producing a simple biosensor. They embedded a measuring electrode and a counter electrode in an insulating base plate so that the ends of the electrodes would be exposed, and disposed a porous material impregnated with enzyme to cover the exposed parts of the electrodes (see Patent Documents 89 and 26). This simplified the user's operating procedure, but biosensors in which this technology is used have yet to become popular as compact devices that allow a blood glucose level to be easily measured at home. This is probably because the problems of manufacturing cost and ease of use have not been solved.
To solve the problems associated with the first-generation biosensors discussed above, research has been conducted into next-generation biosensors in which oxygen or hydrogen peroxide is not used as an electron acceptor, and a dehydrogenase, which is a type of redox enzyme, and a reversible redox electron acceptor are combined (hereinafter referred to as second-generation biosensors) (see Non-Patent Documents 5, 8, and 9, for example).
In 1976, based on the research results of Mindt et al., the Swiss firm of Roche developed the Lactate Analyzer LA640, which makes use of second-generation biosensor technology. This apparatus uses hexacyanoferrate as a soluble electron transfer substance that facilitates the coming and going of electrons between an electrode and lactate dehydrogenase (see Patent Documents 6 and 7).
This apparatus, however, is inconvenient to use, and so was not used in medical clinical applications. Consequently, this apparatus did not enjoy commercial success (see Non-Patent Documents 11 and 14).
In 1978, Nankai et al. disclosed a method for producing an enzyme electrode by combining a redox enzyme and an electron transfer substance, and further specified a plurality of type of electron transfer substances that are useful for disposable biosensors. This indicated that potassium ferricyanide could be applied to disposable biosensors. Potassium ferricyanide is used in nearly all of the currently available biosensors used for measuring blood glucose level (see Patent Document 14).
After the invention by Nankai et al., rapid progress was made in research and development of disposable biosensors between 1980 and 2000. In 1984 ferrocene was employed as an electron transfer substance by Case et al. and Higgins et al., and it was published that a derivative thereof could be used in a practical biosensor by kneading this derivative into an electrode and fixing it (see Patent Document 20 and Non-Patent Documents 10 and 13).
The debut of the first compact amperometric biosensor for measuring blood glucose levels was achieved when the US firm of MediSense produced an enzyme electrode by screen printing (see Patent Documents 19 and 28).
However, these techniques are applications of electronic device technology, and involved the direct application of manufacturing apparatus and methods for manufacturing a printed base plate on which electronic components are mounted. In other words, they involved merely forming a plurality of wires on a non-conductive base plate, and mounting a reaction reagent, and the products were not convenient biosensors that could be used by ordinary consumers. Furthermore, prior to the inventions by Case or Higgins et al., Pace disclosed an ion-selective sensor configuration, and Papakakis disclosed a gas sensor configuration (Patent Documents 16 and 18).
Examples of the specific reasons why the biosensors of by Case or Higgins et al. were inconvenient to use are that measurement required an extremely large amount of blood, and that the measurement accuracy of the sensors was problematic because of the sensor configuration and the manufacturing method.
In view of this, Nankai et al. invented a biosensor in which electrodes were formed mainly from carbon on an insulating base plate, an insulating layer was printed over this, and the electrode surface was accurately restricted, which increased sensor accuracy, and a reaction layer, a spacer, and a cover were also disposed over the electrodes (see, for example, Patent Documents 21, 27, 29, and 371). The use of the spacer and cover dramatically reduced the amount of blood required for measurement. Ultimately, a sensor that required only 2.5 μL of blood was developed as an actual product and is marketed around the world.
This method of forming a liquid sample chamber from a spacer and cover was truly revolutionary, and is employed today in virtually all disposable biosensors used for measuring blood glucose levels. Also, the liquid sample chamber-equipped biosensor perfected by Nankai et al. led to further advances aimed at improving convenience and accuracy (see, for example, Non-Patent Document 15).
In 1993, Yoshioka et al. reported that adhesion of a reagent layer to a carbon electrode on a base plate was increased by treating the carbon electrode surface with an organic solvent. This reagent layer contained an enzyme, an electron transfer substance, and a hydrophilic polymer (see Patent Document 30). With this invention, it was discovered that a problem encountered in the mass production of disposable biosensors, namely, inconsistent quality within or between manufacturing lots, could be minimized. This served as a starting point for giving more importance to measurement accuracy even in mass-produced biosensors.
In 1993, Yoshioka et al. published a biosensor which was equipped with a main electrode system disposed on a base plate and composed of a working electrode and a counter electrode, and which further comprised a reaction layer disposed so as to be in contact with or in the vicinity of the main electrode system. This main electrode system included redox enzyme. With this biosensor, a sub-electrode system is disposed as a reference electrode so as to maintain a distance from the main electrode system. With this biosensor, it can be detected from changes in impedance detected by this reference electrode that enough of a liquid sample has been supplied to the sensor (see Patent Document 32).
With this invention, even diabetes patients who were unaccustomed to using biosensors were able to supply the electrode with the amount of blood needed for measurement with no problem. Consequently, measuring a sufficient quantity of blood should afford better performance, and it is believed that a huge reduction in measurement error should greatly decrease the burden on the patient (see, for example, Non-Patent Document 20).
Still, the following problems were encountered with this invention. A reagent hole is formed by attaching a plate having a through-hole over a base plate on which the electrodes are disposed. However, no liquid sample chamber is employed that would allow a reduction in the amount of blood needed for measurement. Accordingly, pain is inflicted on a diabetes patient, which is a problem in terms of convenience. The fact that the measurement result cannot be obtained until this reaction (the fact that the measurement can be obtained after this reaction) is complete can also be considered an inconvenience to the diabetes patient.
Also, this publication gives examples, albeit very few, of employing a reduction enzyme such as glucose dehydrogenase, rather than just a glucose oxidation enzyme, for a biosensor. An advantage of a dehydrogenase is that it is almost completely unaffected by dissolved oxygen. In particular, glucose dehydrogenase is used in many biosensors manufactured since the late 1990's, and it is also known to have great clinical significance (see, for example, Non-Patent Documents 12, 17, 18, and 19).
In 1995, Kuhn et al. disclosed a biosensor for measuring the hematocrit level of whole blood electrochemically. This biosensor was composed of a working electrode, a counter electrode, and a porous membrane that was disposed spatially separate from these electrodes and contained an electron transfer substance. When whole blood was placed on this porous membrane, a mixture of blood and the electron transfer substance formed. When this mixture reached the electrodes, a current was generated at the electrodes by applying enough potential either to oxidize or to reduce the electron transfer substance. This current was measured, and the hematocrit level was detected from the measurement result (see Patent Document 38).
This invention is a way to handle a hematocrit value, which becomes a problem as an interference substance in making blood into a liquid sample. However, if the viscosity of the blood is extremely high due to excess lipids or the like, there is some doubt as to whether this invention is an effective means. Also, with this invention, it is predicted that it will take longer from the application of a drop of blood until the measurement is complete, and furthermore a relatively large quantity of blood is necessary.
In 1995, White et al. disclosed two measurement devices, as biosensor measurement devices for measuring the amount of a target substance in a biological sample. One was a biosensor measurement device equipped with an algorithm for determining the amount of target substance according to the ambient temperature when a biological sample was in the reaction zone, and the other was a biosensor measurement device with which either a biosensor or a check strip can be inserted (see Patent Documents 39 and 40).
As shown in the above-mentioned inventions of White et al. in 1995, biosensor research and development has been aimed at improving both user convenience and measurement accuracy. This is believed to be attributable to the fact that it was confirmed by the results of a Diabetes Control and Complications Trial published in 1993 that complications tend to occur unless blood glucose level is strictly managed, which led to heightened awareness of simple blood glucose level measurement systems that make use of biosensor systems in clinical medicine. This heightened awareness has led to greater prescription of simple blood glucose level measurement systems to diabetes patients, and even among patients who had never before measured their own blood glucose, there was greater interest in measurement of their own blood (see Non-Patent Document 20).
In 1996, Hill et al. disclosed a strip electrode that involved screen printing. This strip was equipped with a slender support, and this support included first and second conductors that extended along said support. With the biosensor of Hill et al., an active electrode is disposed so as to be in contact with a liquid mixture and the first conductor. An electron transfer substance and an enzyme capable of exerting a catalytic action in the reaction are deposited on this active electrode. A reference electrode is disposed so as to be in contact with the mixture and the second conductor (see, for example, Patent Documents 41, 52, 54, and 60).
To manufacture a biosensor with good measurement accuracy and without variance between products, it is extremely important that the reaction reagent be disposed accurately in a specified location on the biosensor. In view of this, in 1980 Pace et al., and in 1994 Pollman et al., proposed the use of punching technology to form a reagent well including a wall for holding or fixing a reaction reagent at a specific location on a biosensor until the reaction reagent has dried (see Patent Documents 15 and 35).
Nevertheless, such biosensor manufacturing methods generally increase the number of members entailed, and each member needs to be machined very precisely. As a result, the manufacturing process becomes more complicated, and this leads to the problems of lower manufacturing efficiency and higher costs.
In 1990 and 1991, Weetall et al. reported on a well-type biosensor in which electrodes and a reaction reagent were disposed in a well provided to a base plate (see Patent Documents 22 and 24).
This biosensor is a type with which a liquid sample is dropped perpendicularly to the depth direction of the well, making it difficult for the user to use. Also, since the electrodes and the reaction reagent layer have a complicated structure, the cost is higher, so a disadvantage is that this approach is not suited to mass production.
In 1996, Yoshioka et al. proposed the following as a simpler method for manufacturing a biosensor with less variance between products and which was capable of more accurate measurement. In the manufacture of a biosensor comprising an insulating base plate, a working electrode, and a counter electrode, the electrodes were formed in a substantially circular shape, so that when the reaction reagent was applied, a reagent layer could be formed more simply and accurately at the desired locations on the electrodes. The aim of this method is to increase the accuracy of the measurement system itself, and to make the manufacturing processing easier (see Patent Document 42).
In 2001, Winarta et al. reported on a technique for delineating the electrode area where a reaction reagent is disposed on a biosensor. They delineated the electrode area by forming rectangular, square, or circular cuts in a base plate with a carbon dioxide laser, and laminating the resulting plate over a bottom plate. They further held the reaction reagent in an electrode area delineated by the cuts (see Patent Document 67).
However, this method is similar to the above-mentioned methods of Pace et al. and Pollmann et al. in that the number of necessary members is increased and the various members have to be machined very precisely. Therefore, this method also makes the manufacturing process more complicated, the manufacturing efficiency is poor, and the costs are higher.
Other methods that do not involve the use of a well have also been reported as a method for specifying the reaction reagent distribution and the position of the reaction reagent on the biosensor. As an example of this, Bhullar et al. have proposed the following biosensor (see Patent Documents 71, 74, 79, 82, and 85). With this biosensor, a conductive track is formed on a base plate, and an electrode array is formed by this track. A recess is formed near the electrode array on the same base plate. This biosensor also comprises a plate that is disposed on the same base plate, so as to be opposite the surface on which the electrode array and the recess are formed.
With this constitution, the reaction reagent placed on the electrode array spreads over the entire electrode array until it reaches the recess. When the reaction reagent liquid reaches the end of the recess, the interface energy in between the plate and the electrode array falls below the surface tension of the reaction reagent liquid, so the reaction reagent liquid is held on the electrode array. Also, since the reaction reagent liquid is drawn along the edges of the recess, the recess facilitates the diffusion of the reaction reagent liquid on the electrode array.
Compared to the method of Yoshioka et al., the manufacturing method of Bhullar et al. requires an extra machining step of forming a recess in the base plate in order to specify the place where the reaction reagent is to be disposed. Consequently, this method entails more equipment the number of steps increases, and greater machining precision is necessary, so it seems unsuited to the large-scale manufacture of inexpensive biosensors.
As for disposable biosensors, there has not only been technological development of the structure, but measurement methods have also been the focus of considerable technological development. Ikeda et al. in a 1996 report presented a measurement method aimed at higher accuracy. This report deals with a method for quantifying a specific target substance in a liquid sample by using a biosensor having an insulating plate, a working electrode, a counter electrode, and a reaction layer containing an enzyme, in which the liquid sample is drawn into the biosensor, after which the working electrode and the counter electrode are short circuited before the application of voltage for measuring the specific target substance, which eliminates measurement error due to uneven dissolution of the reaction layer, and allows more accurate measurement to be performed (see Patent Document 44).
In 1996, Ikeda et al. disclosed a biosensor comprising a liquid sample detection electrode at a position within a liquid sample chamber and away from a liquid sample intake port. That is, this liquid sample detection electrode is the last to touch the liquid sample of all the electrodes disposed in the liquid sample chamber. Therefore, the fact that a liquid sample is detected by the liquid sample detection electrode means that the liquid sample has diffused over the electrodes (working electrode, counter electrode, etc.) that are important for measurement. With this biosensors, measurement can be commenced after confirming that the liquid sample has thus diffused to the electrodes that are important for measurement (see, for example, Patent Documents 46 and 50).
In 1997, Carter et al. disclosed an electrode strip comprising an electrode support, a reference or counter electrode disposed on the electrode support, a working electrode provided a specific distance away from the reference or counter electrode, a covering layer that defines an enclosed space covering the reference and working electrodes and that has an aperture for receiving a sample into said enclosed space, and a plurality of mesh layers interposed in the enclosed space between the covering layer and the support. This covering layer has a sample application opening provided a specific distance from the electrodes. The working electrode contains an enzyme which can exert a catalytic action on a reaction involving its own substrate, and an electron transfer substance which can transfer electrons between the enzyme-catalyzed reaction and the working electrode. Carter et al. stated that this apparatus reduces the effect of hematocrit on a sensor reading. According to this disclosure, this is accomplished by a combination of the thin layer of the sample solution created by the mesh layers and the position where the reference electrode is disposed with respect to the working electrode (see Patent Document 49).
In 1998, MacAleer et al. disclosed a disposable glucose test strip comprising a plate, a reference electrode, a working electrode, and means for making an electrical connection. The above-mentioned working electrode has a conductive base layer and a covering layer provided covering the conductive base layer. This covering layer is a filler that has both hydrophobic and hydrophilic surface regions forming a network, an enzyme, and an electron transfer substance. The measurement result obtained using this biosensor is not affected by the ambient temperature where the biosensor is used, and exhibits no sensitivity to hematocrit (see Patent Document 53).
In 1998, Henning et al. disclosed a biosensor with which the effect of interfering substances could be reduced. This apparatus generally comprises an electrode for electrochemically measuring the concentration of a target substance in a solution. This apparatus includes peroxidase that is covalently linked to fine particles of carbon and is held in a matrix in a state of close contact with the electrode. With this apparatus, the enzyme/fine particles reduce the effect of known interfering substances of carbon (see Patent Document 55).
In 1998, Charlton et al. disclosed a biosensor having an insulating base plate equipped with an electrode on its surface which reacts with a specimen to produce mobile electrons. This base plate is joined with a cover composed of a deformable material, and has a concave area surrounded by a flat surface so as to form a liquid sample chamber into which a liquid sample can flow. The side of the lid facing the base plate is covered with a polymer material, and the work of this polymer material bonds the lid to the base plate and increases the hydrophilic nature of the capillary space (see Patent Documents 57 and 59).
This biosensor of Charlton et al. works the same as the above-mentioned biosensor equipped with a spacer and cover disclosed by Nankai et al., but its constitution is different, and since there are fewer difficult steps and materials, it is much more likely that this biosensor can be mass-produced inexpensively. However, in the manufacture of this biosensor, selecting the cover material and manufacturing accuracy are extremely important, and difficulties are foreseen in these areas.
In 1998, Pritchard et al. disclosed a biosensor in which the minimum amount of blood sample was only about 9 μL. The main feature of this biosensor is that a working electrode and counter electrode that are substantially the same size and are composed of the same conductive substance are held on a base plate, and these electrodes are covered with an upper cover equipped with a cut-out forming a reagent well. This cut-out exposes a smaller area of the counter electrode than the working electrode. A reagent substantially covers the exposed areas of the working electrode and counter electrode in the reagent well. A mesh that is impregnated with a surfactant covers the reagent well and is affixed to the upper cover (see Patent Document 58). However, even at the time this invention was disclosed, 9 μL was twice the amount of blood as that in the above-mentioned invention of Nankai et al., and this invention was unsuited to reducing the amount of sample needed for measurement.
In 1999, Hoenes et al. reported on a measurement method for simultaneously performing calorimetric and electrochemical measurement, in order to ameliorate the disadvantage of colorimetric detection and reaction. This disadvantage is that while colorimetric detection and reaction is useful at low concentrations of the measurement target substance, measurement is difficult at higher concentrations. The method proposed by Hoenes et al. is as follows. An oxidizing enzyme and a chromogen A that accepts electrons from the enzyme are used in measurement. The chromogen A is reduced to a compound A′, after which a coupling reaction with a substance BX is conducted to form a colored reagent A′B. The concentration of this A′B is measured calorimetrically as an index of the presence or amount of the measurement target substance. Also, since an electrochemically measurable atom group X′ is cleaved off from BX by the coupling reaction, the concentration of this X′ is measured electrochemically as an index of the amount of the measurement target substance. Hoenes et al. thus performed colorimetric and electrochemical measurement of the target substance simultaneously (see Patent Document 61).
Nevertheless, to realize this system, the configuration of the biosensor and measurement apparatus would probably be extremely complicated. And this system can not provide merits (such as measurement accuracy) that offsets this complexity. Also, with an electron transfer system that follows an enzyme reaction, since the reaction reagent composition and the reaction system are complicated, it can be easily imagined that the response speed and reproducibility of the resulting response values will suffer, as will the storage stability of the biosensor.
In 1999, Crismore et al. disclosed a biosensor in which a window is provided to the liquid sample chamber. The main feature of this biosensor is that because a window that is transparent with respect to a colored upper cover is employed for the liquid sample chamber, the user can confirm that enough blood for measurement has been drawn into the liquid sample chamber. Also, if a notch is added to the base plate at the intake port, the intake of the liquid sample can be carried out more smoothly (see Patent Document 62).
But at the time this invention was disclosed, Ikeda et al. had already reported a system in which a measurement device electrochemically detects that a sufficient quantity of liquid sample has been drawn into the liquid sample chamber.
From the 1980's and into the 1990's, screen printing to which printed plate manufacturing technology was applied was used to manufacture biosensors. From about the mid-1990's until the late 1990's, precision machining technology came to be used as the method for manufacturing biosensors to reduce variance between manufacturing lots.
With a biosensor in which the electrodes are formed by screen printing, variance occurs in the surface area of the measuring electrodes due to the bleeding of conductive paste during printing, and this adversely affects the response characteristics. Fujiwara et al. reported on a method of producing a biosensor with good accuracy and by a simple production process instead of using screen printing. This manufacturing method involved forming a metal film over the entire surface of an insulating plate, then dividing the metal film by forming slits, providing a measuring electrode and a counter electrode to which liquid sample is applied and a cover that forms a lead for applying voltage to these electrodes, and covering the measuring electrode and the counter electrode with a reagent layer (see Patent Documents 63 and 69).
With the method of Fujiwara et al., the measuring electrode and the counter electrode are produced by forming slits with a laser or the like in a metal film formed by through vapor deposition, sputtering, or by bonding a metal foil over an insulating plate. Accordingly, there is no bleeding of the print as in screen printing, and the surface area of the electrodes can be accurately defined. Therefore, there is less variance in response characteristics among sensors, and a biosensor with good precision can be obtained.
Advantages to laser machining are that the manufacturing steps and the required equipment are not complicated, manufacturing reproducibility is good, and so forth, and this machining is extremely useful in the manufacture of biosensors. The working of a metal film on an insulator is a technique applied in many fields other than biosensors, and its usefulness has been corroborated in numerous reports (see, for example, Patent Documents 25, 43, 45, 47, and 56).
Also, from 2000 onward, there have been reports of applying precision machining by laser in the field of biosensor manufacturing technology. These include a method for forming an electrode by laser machining of a metal layer on an insulating plate (see Patent Documents 81, 84, and 87).
There have been many reports up to now concerning the development of systems for measuring with greater accuracy. These relate to the structure of a biosensor, or relate to a manufacturing method, or relate to a measurement method. All current biosensors, however, still have various drawbacks.
One of these drawbacks is interference with the biosensor reading brought about by other substances which are present in a liquid sample and can oxidize at the same potential as the measurement target. These are referred to as interference substances in this Specification, and typical examples include ascorbic acid, uric acid, and acetaminophen. When these and other interference substances oxidize, the current produced by their oxidation is on top of the current intrinsic to the measurement target, and cannot be distinguished from the current intrinsic to the measurement target, so the concentration thereof is estimated too high. As a result, accuracy drops in the quantification of the measurement target substance in the liquid sample.
The 2001 report by Ikeda et al. proposed a method for reducing the effect of interference substances. The biosensor of Ikeda et al. had a working electrode, a counter electrode and a reagent layer, and had a third electrode not in contact with the reagent layer, in order to reduce the decrease in measurement accuracy caused by interference substances. Ikeda et al. aimed to reduce the effect of interference substances by using this third electrode as an interference substance detection electrode (see Patent Document 65).
Another drawback to biosensors in which the liquid sample is blood is interference caused by red blood cells (hematocrit effect). This interference tends to result in an apparently high reaction rate with respect to a low hematocrit level, or conversely, to result in an apparently low reaction rate with respect to a high hematocrit level.
In 1984, Vogel et al. reported on a biosensor in which glass fibers were deposited as a technique for reducing the hematocrit effect. This biosensor separated or removed red blood cell from blood by passage of blood through glass fibers. Whereas red blood cells were usually separated or removed by centrifuging, with this method, red blood cells could be separated without pretreatment, but the deposition of glass fibers is expected to drive up material costs and manufacturing costs, and to adversely affect manufacturing precision. In addition, since blood is passed through fibers whose diameter is sufficient to separate red blood cells, the rate at which the sample is drawn in is low, and this probably makes the product less convenient to use.
Also, in 2001, Winarta et al. reported on the composition of a reaction reagent that reacts with a measurement target substance. In this report, a polymer stabilizer, a binder, and a surfactant are added along with an enzyme and an electron transfer substance to the reaction reagent (see Patent Document 66).
However, adding these substances to the reaction reagent causes problems in that it makes the reagent preparation process more complicated, makes the reagent layer formation process more complicated by increasing the reaction reagent viscosity, adversely affects the drying process, leads to a decrease in the reaction reagent reaction efficiency, decreases the long-term storage stability of the reaction reagent, and so forth.
In addition to the above drawbacks, other drawbacks to prior art are that its linear range is limited, and that a relatively large quantity of sample is needed. Furthermore, these apparatus take a comparatively long time until the steady response prior to obtaining a reading becomes obvious. Each of these drawbacks individually, or combined with one or more other drawbacks, can be a source of an erroneous measurement reading in analysis. In preliminary tests conducted by the inventors of the present invention, prior art that claimed to reduce the effect of hematocrit on a glucose reading was limited to a low glucose concentration, and indicated that there was an effect only at a low glucose concentration. At the higher blood glucose levels that actual diabetes patients can display, the desired effect is not exhibited.
Technology aimed at reducing the volume of liquid sample necessary for measurement in an attempt to make the product more convenient to use for the user has been reported up to now. For instance, there have been various studies into sensor structure, and particularly the structure of the liquid sample chamber, or electrode layout, reaction reagent composition and additives, measurement methods, and so forth (see, for example, Patent Documents 68, 70, 72, 75, 77, and 78).
Even if the minimum amount of liquid sample needed is reduced, though, the puncturing that causes the user pain is still not eliminated. Also, if the required amount is reduced so much that the space inside the liquid sample chamber is filled with the reagent layer, this will likely have a detrimental effect on how the liquid sample is drawn in. Also, if the liquid sample chamber is merely made smaller to reduce the required amount, then the electrode surface area and the reagent layer will also become smaller at the same time, so the response value obtained during measurement will be smaller, and as a result, the signal to noise ratio (S/N ratio) will deteriorate and measurement accuracy will suffer.
Recent developments of biosensors have often attempted to enhance both convenience and accuracy. Of these attempts, particular emphasis has been placed on shortening the measurement time and reducing the amount of liquid sample required for measurement.
In measurement with a biosensor, the reaction reagent disposed in a dry state must be redissolved by the blood or other liquid sample. Therefore, to shorten the measurement time, fast reactivity is necessary, and higher resolubility of the reaction reagent can be considered an advantage. A problem with a reaction reagent with excellent resolubility, however, is that the reagent ends up being carried away in the intake direction when the liquid sample is drawn in. If the reaction reagent is carried away, this decreases the reaction reagent concentration in the reaction area, so the measured value will be lower than the true value. Also, if the intake rate should change due to action on the part of the user or the hematocrit value, the reaction reagent concentration in the reaction area will also change, so measurement reproducibility is greatly diminished.
This problem has become much more critical as biosensor development has advanced in recent years. Methods in which a reaction reagent is fixed on an electrode using a polymer or a sol-gel matrix have been reported as a method for forming a reaction reagent layer that is not carried away (see Patent Documents 31, 33, 34, 36, 48, and 51).
These reports involve the use of an electron transfer substance that is non-leachable and non-diffusible. However, when the leaching or diffusion of a reaction reagent is suppressed by adding a polymer to the reaction reagent, or by fixing with covalent bonds, or by supporting the reaction reagent in a sol-gel matrix, it becomes impossible to conduct a fast reaction within a limited amount of time. This is contrary to the biosensor development trend of recent years, which is to try and shorten measurement time.
In 2004, Bhullar et al. reported an attempt to achieve a suitably controlled flow of liquid sample by forming a microstructure by working the basal part of an intake port distal end (see Patent Documents 80 and 83).
In these reports, the flow of liquid sample is adjusted by capillary force during liquid sample intake. However, even though the direction in which the liquid sample flows can be controlled, the intake rate cannot be controlled, and it is difficult to prevent the reaction reagent from being carried away. Also, in these reports, the base plate is worked by laser ablation, but this requires sophisticated machining technology and considerable equipment to form a microstructure such as that proposed. Putting this technique into actual practice is probably going to be met with many difficulties.
An example of working a base plate into a groove shape is the report of Say et al. Their report is aimed at a biosensor implanted in subcutaneous interstitial tissue of the patient, and the base plate is worked into a groove shape, and a conductive substance is placed in the groove thus formed, thereby forming an embedded electrode (see Patent Documents 64, 73, and 86).
The basic structure of a biosensor in which a reaction reagent is disposed on this embedded electrode is not that different from the structure of a conventional embedded electrode, so this does not seem to solve the problem of controlling the extent to which the reaction reagent is carried away by liquid sample intake. Also, with biosensors that are expected to see use under a variety of conditions, the flexibility of the base plate in these biosensors is predicted to be affected by the usage environment, which will likely pose problems in practical application.
In the embodiments given in the 2003 report of Feldman et al., a groove is formed in a biosensor by subjecting the base plate to embossing (see Patent Document 76).
In this case, the opposite side of the base plate on which the groove is formed by embossing ends up being worked into a peaked shape. This portion is the portion that comes into direct contact with user when the user mounts the biosensor in the measurement device, so the product is obviously less comfortable to use if the shape is peaked. Also, if the peaked portion should be deformed during transport or when touched by the user, the groove shape and volume may be changed, and this could have a serious adverse affect on measurement accuracy. Furthermore, embossing requires that the material be carefully selected as well. For the above reasons, it is believed that there are still problems with this technique if the convenience and measurement accuracy demanded of a biosensor are both to be satisfied.
In 2006, Huang reported on an embedded electrode type of biosensor in which a liquid sample chamber was formed by providing a recess to the base plate (see Patent Document 88).
With this biosensor, the electrode is embedded in a hole in the base plate, and a liquid sample chamber is not formed by laminating a spacer over the base plate, but rather by providing a recess to the base plate. This reduces the number of parts needed for manufacturing a biosensor, but high precision is required of the recess formation because the shape of recess defines the shape and volume of the liquid sample chamber. Also, since the basic structure of the liquid sample chamber is not that different from a conventional design, this does not seem to solve the problem of controlling the extent to which the reaction reagent is carried away by liquid sample intake.
These biosensors are sensors that make use of the molecular recognition capability of biological materials such as microbes, enzymes, antibodies, and nucleic acids. Specifically, they make use of various kinds of biochemical reactions that occur when a biological material recognizes the specific component being targeted, such as the consumption of oxygen by the respiration of microorganisms, an enzyme reaction, coloration by a coloring reagent, or the like. Research and application of biosensors utilizing enzymes have been particularly active, and these have seen practical use in the medical field and food industry. Of these, biosensors used in the medical field are used by diabetes patients to measure their own blood glucose levels, and are sold around the world.
The following is a list of US patent documents, Japanese patent documents, and non-patent documents cited in this Specification.
Patent Document 1: U.S. Pat. No. 2,913,386, November 1959, Clark
Patent Document 2: U.S. Pat. No. 3,539,455, November 1970, Clark
Patent Document 3: U.S. Pat. No. 3,542,662, November 1970, Hicks et al.
Patent Document 4: U.S. Pat. No. 3,770,607, November 1973, William
Patent Document 5: U.S. Pat. No. 3,788,950, January 1974, Hicks et al.
Patent Document 6: U.S. Pat. No. 3,836,003, September 1974, Mindt et al.
Patent Document 7: U.S. Pat. No. 3,838,033, September 1974, Mindt et al.
Patent Document 8: U.S. Pat. No. 3,948,745, April 1976, Guilbault et al.
Patent Document 9: U.S. Pat. No. 3,979,274, September 1976, Newman et al.
Patent Document 10: U.S. Pat. No. 4,073,713, February 1978, Newman et al.
Patent Document 11: U.S. Pat. No. 4,085,009, April 1978, Pace
Patent Document 12: U.S. Pat. No. 4,092,233, May 1978, Clemens et al.
Patent Document 13: U.S. Pat. No. 4,100,029, July 1978, Prosper
Patent Document 14: U.S. Pat. No. 4,224,125, September 1978, Nankai et al.
Patent Document 15: U.S. Pat. No. 4,225,410, September 1980, Pace et al.
Patent Document 16: U.S. Pat. No. 4,454,007, June 1984, Pace
Patent Document 17: U.S. Pat. No. 4,477,575, October 1984, Vogel et al.
Patent Document 18: U.S. Pat. No. 4,534,356, August 1985, Papadakis
Patent Document 19: U.S. Pat. No. 4,545,382, October 1985, Higgins et al.
Patent Document 20: U.S. Pat. No. 4,711,245, December 1987, Higgins et al.
Patent Document 21: U.S. Pat. No. 4,897,173, January 1990, Nankai et al.
Patent Document 22: U.S. Pat. No. 4,963,245, October 1990, Weetall et al.
Patent Document 23: U.S. Pat. No. 4,975,175, December 1990, Karube et al.
Patent Document 24: U.S. Pat. No. 5,066,372, November 1991, Weetall et al.
Patent Document 25: U.S. Pat. No. 5,104,480, April 1992, Wojnarowski et al.
Patent Document 26: U.S. Pat. No. 5,120,420, June 1992, Nankai et al.
Patent Document 27: U.S. Pat. No. 5,120,420, June 1992, Nankai et al.
Patent Document 28: U.S. Pat. No. 5,126,034, June 1992, Carter et al.
Patent Document 29: U.S. Pat. No. 5,185,256, February 1993, Nankai et al.
Patent Document 30: U.S. Pat. No. 5,229,282, July 1993, Yoshioka et al.
Patent Document 31: U.S. Pat. No. 5,262,035, November 1993, Gregg et al.
Patent Document 32: U.S. Pat. No. 5,264,103, November 1993, Yoshioka et al.
Patent Document 33: U.S. Pat. No. 5,264,104, November 1993, Gregg et al.
Patent Document 34: U.S. Pat. No. 5,264,105, November 1993, Gregg et al.
Patent Document 35: U.S. Pat. No. 5,288,636, February 1994, Pollmann et al.
Patent Document 36: U.S. Pat. No. 5,320,725, June 1994, Gregg et al.
Patent Document 37: U.S. Pat. No. 5,320,732, June 1994, Nankai et al.
Patent Document 38: U.S. Pat. No. 5,385,846, January 1995, Kuhn et al.
Patent Document 39: U.S. Pat. No. 5,405,511, April 1995, White et al.
Patent Document 40: U.S. Pat. No. 5,438,271, August 1995, White et al.
Patent Document 41: U.S. Pat. No. 5,509,410, April 1996, Hill et al.
Patent Document 42: U.S. Pat. No. 5,512,159, April 1996, Yoshioka et al.
Patent Document 43: U.S. Pat. No. 5,512,489, April 1996, Girault et al.
Patent Document 44: U.S. Pat. No. 5,565,085, October 1996, Ikeda et al.
Patent Document 45: U.S. Pat. No. 5,576,073, November 1996, Kickelhain et al.
Patent Document 46: U.S. Pat. No. 5,582,697, December 1996, Ikeda et al.
Patent Document 47: U.S. Pat. No. 5,593,739, January 1997, Kickelhain et al.
Patent Document 48: U.S. Pat. No. 5,593,852, January 1997, Heller et al.
Patent Document 49: U.S. Pat. No. 5,628,890, May 1997, Carter et al.
Patent Document 50: U.S. Pat. No. 5,650,062, July 1997, Ikeda et al.
Patent Document 51: U.S. Pat. No. 5,665,222, September 1997, Heller et al.
Patent Document 52: U.S. Pat. No. 5,682,884, November 1997, Hill et al.
Patent Document 53: U.S. Pat. No. 5,708,247, January 1998, MacAleer et al.
Patent Document 54: U.S. Pat. No. 5,727,548, March 1998, Hill et al.
Patent Document 55: U.S. Pat. No. 5,755,953, May 1998, Henning et al.
Patent Document 56: U.S. Pat. No. 5,758,398, June 1998, Rijnbeek et al.
Patent Document 57: U.S. Pat. No. 5,759,364, June 1998, Charlton et al.
Patent Document 58: U.S. Pat. No. 5,762,770, June 1998, Pritchard et al.
Patent Document 59: U.S. Pat. No. 5,798,031, August 1998, Charlton et al.
Patent Document 60: U.S. Pat. No. 5,820,551, October 1998, Hill et al.
Patent Document 61: U.S. Pat. No. 5,858,691, January 1999, Hoenes et al.
Patent Document 62: U.S. Pat. No. 5,997,817, December 1999, Crismore et al.
Patent Document 63: U.S. Pat. No. 6,004,441, December 1999, Fujiwara et al.
Patent Document 64: U.S. Pat. No. 6,103,033, August 2000, Say et al.
Patent Document 65: U.S. Pat. No. 6,212,417, April 2001, Ikeda et al.
Patent Document 66: U.S. Pat. No. 6,258,229, July 2001, Winarta et al.
Patent Document 67: U.S. Pat. No. 6,287,451, September 2001, Winarta et al.
Patent Document 68: U.S. Pat. No. 6,299,757, October 2001, Feldman et al.
Patent Document 69: U.S. Pat. No. 6,309,526 October 2001 Fujiwara et al.
Patent Document 70: U.S. Pat. No. 6,338,790, January 2002, Feldman et al.
Patent Document 71: U.S. Pat. No. 6,447,657, September 2002, Bhullar et al.
Patent Document 72: U.S. Pat. No. 6,461,496, October 2002, Feldman et al.
Patent Document 73: U.S. Pat. No. 6,484,046, November 2002, Say et al.
Patent Document 74: U.S. Pat. No. 6,540,890, April 2003, Bhullar et al.
Patent Document 75: U.S. Pat. No. 6,576,101, June 2003, Heller et al.
Patent Document 76: U.S. Pat. No. 6,592,745, July 2003, Feldman et al.
Patent Document 77: U.S. Pat. No. 6,616,819, September 2003, Liamos et al.
Patent Document 78: U.S. Pat. No. 6,618,934, September 2003, Feldman
Patent Document 79: U.S. Pat. No. 6,645,359, November 2003, Bhullar et al.
Patent Document 80: U.S. Pat. No. 6,755,949, June 2004, Bhullar et al.
Patent Document 81: U.S. Pat. No. 6,767,440, July 2004, Bhullar et al.
Patent Document 82: U.S. Pat. No. 6,814,843, November 2004, Bhullar et al.
Patent Document 83: U.S. Pat. No. 6,814,844, November 2004, Bhullar et al.
Patent Document 84: U.S. Pat. No. 6,866,758, March 2005, Bhullar et al.
Patent Document 85: U.S. Pat. No. 6,911,621, June 2005, Bhullar et al.
Patent Document 86: U.S. Pat. No. 7,003,340, February 2006, Say et al.
Patent Document 87: U.S. Pat. No. 7,073,246, July 2006, Bhullar et al.
Patent Document 88: U.S. Laid-Open Patent Application No. 2006/0175199, August 2006, Huang
Patent Document 89: Japanese Patent Publication No. H4-6907 (Japanese Laid-Open Patent Application No. S59-166852), February 1992, Kawaguri et al.
Non-Patent Document 1: L. C. Clark, Jr., “Monitor and control of blood and tissue oxygenation,” Trans. Am. Soc. Artif. Intern. Organs, 1956 2, 41.
Non-Patent Document 2: L. C. Clark, Jr., “Electrode system for continuous monitoring in cardiovascular surgery,” Ann. NY Acad. Sci,. 1962 102:29/45.
Non-Patent Document 3: Voss D O., “A new oxygen electrode model for the polarographic assay of cellular and mitochondrial respiration,” Anal Biochem. Sep. 6, 1963; 211-222.
Non-Patent Document 4: Updike, S J. et al., “The enzyme electrode,” Nature Jun. 3, 1967; 214(5092): 986-988.
Non-Patent Document 5: Fujihira et al., “Reversible redox titrations of cytochrome c and cytochrome c oxidase using detergent solubilized electrochemically generated mediator-titrations,” Biochemical and Biophysical Research Communications 1974 61(2): 538-543.
Non-Patent Document 6: Divies, C., “Remarks on ethanol oxidation by an ‘Acetobacter xylinum’ microbial electrode,” Ann Microbiol (Paris), 1975 February-March; 126(2): 175-86. (Original text in French, and this title is an English translation by the author.)
Non-Patent Document 7: L. C. Thomas et al., “Voltammetric measurement of reduced Nicotinamide-Ademine Nucleotides and application to amperometric measurements of enzyme reactions,” Analytica Chemica Acta 1975, 78, 271-276.
Non-Patent Document 8: Yagi et al., “A new assay method for hydrogenase based on an enzymic electrode reaction. The enzymic electric cell method,” J Biochem. 1975 September; 78 (3): 443-454.
Non-Patent Document 9: Johnson, J M. et al., “Metal complex as mediator-titrants for electrochemical studies of biological systems,” Anal. Biochem. 1983 August; 133 (1): 186-189.
Non-Patent Document 10: Cass et al., “Ferrocene-mediated enzyme electrode for amperometric determination of glucose,” Ana. Chem. 1984 April; 56 (4): 667-671.
Non-Patent Document 11: Denis, C. et al., “Use of the LA 640 for a simple method of measuring the concentration of muscle lactate” J Physiol. 1985; 80(3): 168-172.
Non-Patent Document 12: D'Costa, E. J. et al., “Quinoprotein glucose dehydrogenase and its applications in an amperometric glucose sensor,” Biosensors 1986 2: 71-87.
Non-Patent Document 13: Cass et al., “Biosensors: Fundamentals and Applications,” Oxford University Press, 1987 & 1989.
Non-Patent Document 14: Denis, C. et al., “Effects of endurance training on hyperammonaemia during a 45-min constant exercise intensity,” Eur J Appl Physiol Occup Physiol. 1989; 59 (4): 268-272.
Non-Patent Document 15: Morris, N. A. et al., “Electrochemical capillary fill device for the analysis of glucose incorporating glucose oxidase and ruthenium(III) hexamine as mediator,” Electroanalysis 1992 4: 1-9.
Non-Patent Document 16: The Diabetes Control and Complications Trial Research Group, “The Effect of Intensive Treatment of Diabetes on the Development and Progression of Long-Term Complications in Insulin-Dependent Diabetes Mellitus,” New England Journal of Medicine, Sep. 30, 1993 329 (14) 977-986.
Non-Patent Document 17: Loughran, M. G. et al., “Amperometric detection of histamine at a quinoprotein dehydrogenase enzyme electrode,” Biosensors and Bioelectronics, 1995 10: 569-576.
Non-Patent Document 18: Loughran, M. G. et al., “Ammonium requirement and stability of methanol dehydrogenase TTF. TCNQ electrodes,” The Analyst, 1996 121: 1711-1715.
Non-Patent Document 19: Loughran, M. G. et al., “Development of apyrroloquinoline quinone (PQQ) mediated glucose oxidase enzyme electrode for detection of glucose in fruit juice,” Electroanalysis, 1996 8 (10) 870-875.
Non-Patent Document 20: Newman, J D. et al., “Home blood glucose biosensors: a commercial perspective,” Biosens Bioelectron. Jun. 15, 2005; 20 (12): 2435-2453.